Isolation of a human hepatic 60 kDa aspartylglucosaminidase consisting of three non-identical polypeptides

Biochemical characterization and comparison of aspartylglucosaminidases secreted in venom of the parasitoid wasps Asobara tabida and Leptopilina heterotoma

Aspartylglucosaminidase (AGA) is a low-abundance intracellular enzyme that plays a key role in the last stage of glycoproteins degradation, and whose deficiency leads to human aspartylglucosaminuria, a lysosomal storage disease. Surprisingly, high amounts of AGA-like proteins are secreted in the venom of two phylogenetically distant hymenopteran parasitoid wasp species, Asobara tabida (Braconidae) and Leptopilina heterotoma (Cynipidae). These venom AGAs have a similar domain organization as mammalian AGAs. They share with them key residues for autocatalysis and activity, and the mature α- and β-subunits also form an (αβ)2 structure in solution. Interestingly, only one of these AGAs subunits (α for AtAGA and β for LhAGA) is glycosylated instead of the two subunits for lysosomal human AGA (hAGA), and these glycosylations are partially resistant to PGNase F treatment. The two venom AGAs are secreted as fully activated enzymes, they have a similar aspartylglucosaminidase activity and are both also efficient asparaginases. Once AGAs are injected into the larvae of the Drosophila melanogaster host, the asparaginase activity may play a role in modulating their physiology. Altogether, our data provide new elements for a better understanding of the secretion and the role of venom AGAs as virulence factors in the parasitoid wasps' success.

Large-scale purification and preliminary x-ray diffraction studies of human aspartylglucosaminidase

Aspartylglucosaminidase (AGA) is a lysosomal asparaginase that takes part in the ordered degradation of glycoproteins and a deficiency of which results in a lysosomal accumulation disease aspartylglucosaminuria in human. The mature enzyme consists of 24-kDa and 17-kDa subunits, which are both heterogeneously glycosylated. Activation of the enzyme from a single precursor polypeptide into two subunits is accomplished in the endoplasmic reticulum (ER). The relative lack of this proteolytic capacity in several tested high-producing expression systems has complicated the production of active recombinant enzyme in high quantities, which would be an alternative for purification of this molecule for crystallization. Consequently, the AGA enzyme has to be purified directly from cellular or tissue sources for crystallographic analysis. Here we describe a large-scale purification method to produce milligram amounts of homogeneous AGA from human leukocytes. The purified AGA enzyme represents a heterogeneous pool of molecules not only due to glycosylation, but also heterogeneity at the polypeptide level, as demonstrated here. We were able to isolate a homogeneous peptide pool that was successfully crystallized and preliminary X-ray data collected from the crystals. The crystals diffract well to 2.0 angstroms and are thus suitable for determination of the crystal structure of AGA.

Human leucocyte glycosylasparaginase is an alpha/beta-heterodimer of 19 kDa alpha-subunit and 17 and 18 kDa beta-subunit

Human lysosomal glycosylasparaginase (AGA; EC 3.5.1.26) consists of two glycosylated subunits, alpha and beta. Treatment with 3% SDS at 45 degrees C as part of a new purification scheme did not affect enzyme activity, but the alpha-subunit migrated an apparent 19 kDa peptide on SDS/PAGE instead of as a 24 kDa peptide, as observed without this SDS treatment. The N-terminal sequence was similar to that of the 24 kDa form, and, after reversed-phase h.p.l.c., the 19 kDa form was transformed to an apparent 24 kDa peptide on SDS/PAGE, indicating that their primary structures were identical. As the molecular mass of the alpha-subunit deduced from its cDNA was 19.5 kDa, the variation might be due to incomplete SDS coating of the 24 kDa form. This was confirmed by the tendency of the 24 kDa variant to polymerize even in the presence of SDS. The molecular mass of the beta-subunit was 17 and 18 kDa in accordance with previous reports. Chemical cross-linking with 1-ethyl-3-(3-dimethylaminopropyl)carbodi-imide resulted in the appearance of a 38 kDa peptide on SDS/PAGE which reacted with both the subunit-specific antisera on Western-blot analysis. On SDS/PAGE at pH 10.2 the active enzyme migrated as an apparent 43 kDa peptide. These results indicate that native human glycosylasparaginase is a heterodimer.

Dissection of the molecular consequences of a double mutation causing a human lysosomal disease

Aspartylglucosaminidase (AGA) is a lysosomal enzyme, the deficiency in which leads to human storage disease aspartylglucosaminuria (AGU). AGUFin is the most common AGU mutation in the world and is found in 98% of AGU alleles in Finland, where the population displays enrichment of the disease allele. The AGUFin allele actually contains a double mutation, both individual mutations resulting in amino acid substitutions: Arg-161-->Gln and Cys-163-->Ser. The separate consequences of these two amino acid substitutions for the intracellular processing of the AGA polypeptides were analyzed using a stable expression of mutant polypeptides in Chinese hamster ovary (CHO) cells. The synthesized polypeptides were monitored by metabolic labeling, followed by immunoprecipitation, immunofluorescence, and immunoelectron microscopy. The Arg-161-->Gln substitution did not affect the intracellular processing or transport of AGA and the fully active enzyme was correctly targeted to lysosomes. The Cys-163-->Ser substitution prevented the early proteolytic cleavage required for the activation of the precursor AGA polypeptide and the inactive enzyme was accumulated in the endoplasmic reticulum (ER). The precursors of the translation products of the AGUFin double mutant and the Cys-163-->Ser mutant were also observed in the culture medium. When cells expressing the normal AGA or AGUFin double mutation were treated with DTT to prevent the formation of disulfide bonds, both normal and mutated AGA polypeptides remained in the inactive precursor form and were not secreted into the medium. These results indicate that correct initial folding is essential for the proteolytic activation of AGA.(ABSTRACT TRUNCATED AT 250 WORDS)

Dissection of the molecular pathology of aspartylglucosaminuria provides the basis for DNA diagnostics and future therapeutic interventions

Aspartylglucosaminuria (AGU) is exceptional among lysosomal storage diseases since it represents the only known amidase deficiency in man, being caused by an inadequate function of aspartylglucosaminidase (AGA, E.C. 3.5.1.26.). This amidase is essential in one of the final steps in the ordered breakdown of glycoproteins since it cleaves Asn from the residual N-acetylglucosamines (for reviews see 1, 2). The deficiency of the enzyme activity results in the typical lysosomal accumulation of the abnormal degradation products (mainly aspartylglucosamine, 2-acetamido-1-beta-L-aspartamido-1,2-dideoxyglucose) in patients' cells and tissues. The diagnosis of AGU has so far been based on the detection of abnormal metabolites in urine and decreased enzyme activity in the cultured fibroblasts or isolated lymphocytes. Prenatal diagnosis has been possible by demonstrating the deficient enzyme activity of amniocytes or chorion villus biopsies. Identification of carriers has been difficult and unreliable due to the high individual variation in AGA activity and prerequisite for isolated blood lymphocytes. During the past few years we have purified the human enzyme into homogeneity, isolated the full length cDNA and characterized the majority of AGU mutations in this cDNA. This work facilitated the development of a reliable DNA diagnostic test suitable also for large scale carrier screening. The molecular pathology of the most common AGU mutation was unravelled, this being a prerequisite for the oncoming developments for therapy. Although AGU is a relatively rare disease, characterization of the AGU mutations and their cellular consequences have revealed highly interesting new phenomena in the biosynthesis of this lysosomal enzyme, some of which carry general biological significance.(ABSTRACT TRUNCATED AT 250 WORDS)

Purification and structure of human liver aspartylglucosaminidase

We have recently diagnosed aspartylglucosaminuria (AGU) in four members of a Canadian family. AGU is a lysosomal storage disease in which asparagine-linked glycopeptides accumulate to particularly high concentrations in liver, spleen and thyroid of affected individuals. A lesser accumulation of these glycopeptides is seen in the kidney and brain, and they are also excreted in the urine. The altered metabolism in AGU results from a deficiency of the enzyme aspartylglucosaminidase (1-aspartamido-beta-N-acetylglucosamine amidohydrolase), which hydrolyses the asparagine to N-acetylglucosamine linkages of glycoproteins and glycopeptides. We have used human liver as a source of material for the purification of aspartylglucosaminidase. The enzyme has been purified to homogeneity by using heat treatment, (NH4)2SO4 fractionation, and chromatography on concanavalin A-Sepharose, DEAE-Sepharose, sulphopropyl-Sephadex, hydroxyapatite, DEAE-cellulose and Sephadex G-100. Enzyme activity was followed by measuring colorimetrically the N-acetylglucosamine released from aspartylglucosamine at 56 degrees C. The purified enzyme protein ran at a 'native' molecular mass of 56 kDa in SDS/12.5%-PAGE gels, and the enzyme activity could be quantitatively recovered at this molecular mass by using gel slices as enzyme source in the assay. After denaturation by boiling in SDS the 56 kDa protein was lost with the corresponding appearance of polypeptides alpha,beta and beta 1, lacking enzyme activity, at 24.6, 18.4 and 17.4 kDa respectively. Treatment of heat-denatured enzyme with N-glycosidase F resulted in the following decreases in molecular mass; 24.6 to 23 kDa and 18.4 and 17.4 to 15.8 kDa. These studies indicate that human liver aspartylglucosaminidase is composed of two non-identical polypeptides, each of which is glycosylated. The N-termini of alpha,beta and beta 1 were directly accessible for sequencing, and the first 21, 26 and 22 amino acids respectively were identified.

Human aspartylglucosaminidase. A biochemical and immunocytochemical characterization of the enzyme in normal and aspartylglucosaminuria fibroblasts

Aspartylglucosaminidase (AGA, EC 3.5.1.26) is an essential enzyme in the degradation of asparagine-linked glycoproteins. In man, deficient activity of this enzyme leads to aspartylglucosaminuria (AGU), a recessively inherited lysosomal storage disease. Here we used affinity-purified polyclonal antibodies against the native AGA and its denatured subunits to establish the molecular structure and intracellular location of the enzyme in normal and AGU fibroblasts. Inactivation of the enzyme was found to coincide with the dissociation of the heterodimeric enzyme complex into subunits. Although the subunits were not linked by covalent forces, the intrapolypeptide disulphide bridges were found to be essential for the normal function of AGA. AGA was localized into lysosomes in control fibroblasts by both immunofluorescence microscopy and immuno-electron microscopy, whereas in AGU cells the location of antigen was different, suggesting that, owing to the mutation, a missing disulphide bridge, most of the enzyme molecules get retarded in the cis-Golgi region and most probably face intracellular degradation.

Comparison of liver glycosylasparaginases from six vertebrates

Structural and physical properties of glycosylasparaginase (EC 3.5.1.26) from the livers of human, pig, cow, rat, mouse and chicken were compared. The enzyme in all species had a common basic structure of two N-glycosylated subunits of about 24 (alpha) and 20 (beta) kDa joined by non-covalent forces. Subunit-specific antisera against the rat glycosylasparaginase bound specifically and sensitively to the corresponding subunits from all species. Identity of 80% of the amino acids was found between the N-terminal sequences of corresponding pig and rat glycosylasparaginase alpha- and beta-subunits and the deduced sequence from a human glycosylasparaginase cDNA [Fisher, Tollersrud & Aronson (1990) FEBS Lett. 269, 440-444]. The beta-subunit from all three species has an N-terminal threonine reported to be involved in the reaction mechanism for the human enzyme [Kaartinen, Williams, Tomich, Yates, Hood & Mononen (1991) J. Biol. Chem. 266, 5860-5869]. The native enzyme appeared as a heterodimer among the mammals, whereas the chicken enzyme had a greater molecular mass and is probably either a tetramer or a heterodimer bound to an unrelated peptide(s). All glycosylasparaginases were thermostable, requiring temperatures between 65 degrees C and 80 degrees C to be irreversibly inactivated. In addition, they were unusually stable at high pH and remained active in the presence of SDS except at low pH. The pH maximum was between 5.5 and 6 except for the rat and mouse enzymes which had a broad maximum between pH 7 and 8. A number of other properties were observed which also distinguish the enzyme from individual and closely related species.

In vitro mutagenesis helps to unravel the biological consequences of aspartylglucosaminuria mutation

Aspartylglucosaminuria (AGU) is a lysosomal storage disease resulting in severe mental retardation. We have recently reported that mutations in the aspartylglucosaminidase (AGA) locus are responsible for this disease. About 90% of reported AGU cases are found in Finland, and we have shown that the vast majority (98%) of AGU alleles in this isolated population contain two point mutations located 5 bp apart. We expressed these Arg161----Gln and Cys163----Ser mutations separately in vitro and demonstrated that deficient enzyme activity is caused by the Cys163----Ser mutation, whereas the Arg161----Gln substitution represents a rare polymorphism. Further analyses of in vitro expressed AGA proteins and the enzyme purified from an AGU patient revealed that Cys163 participates in and S-S bridge. The absence of this covalent cross-link in the mutated protein most probably results in disturbed folding of the polypeptide chain and a consequent decrease in its intracellular stability.

Genomic structure of human lysosomal glycosylasparaginase

The gene structure of the human lysosomal enzyme glycosylasparaginase was determined. The gene spans 13 kb and consists of 9 exons. Both 5' and 3' untranslated regions of the gene are uninterrupted by introns. A number of transcriptional elements were identified in the 5' upstream sequence that includes two putative CAAT boxes followed by TATA-like sequences together with two AP-2 binding sites and one for Spl. A 100 bp CpG island and several ETF binding sites were also found. Additional AP-2 and Sp1 binding sites are present in the first intron. Two polyadenylation sites are present and appear to be functional. The major known glycosylasparaginase gene defect G488----C, which causes the lysosomal storage disease aspartylglycosaminuria (AGU) in Finland, is located in exon 4. Exon 5 encodes the post-translational cleavage site for the formation of the mature alpha/beta subunits of the enzyme as well as a recently proposed active site threonine, Thr206.

Glycoasparaginase in human urine

Glycoasparaginase was purified 15,000-fold from human urine. The enzyme is a tetrameric protein of 86 kDa, composed of two heavy chains (25 kDa) and two light chains (18 kDa). Its structure and properties are very similar to those of human leukocyte glycoasparaginase. Glycoasparaginase activity is totally absent from urine of aspartylglycosaminuria patients.

Human leucocyte aspartylglucosaminidase. Evidence for two different subunits in a more complex native structure

Human leucocyte aspartylglucosaminidase (AGA: 1-aspartamido-beta-N-acetylglucosamine amidohydrolase, EC 3.5.1.26) was purified to homogeneity by using affinity chromatography, gel filtration, chromatofocusing and reverse-phase h.p.l.c. As shown by SDS/PAGE, the homogeneous purified enzyme preparation consists of four polypeptide chains with molecular masses of 25, 24, 18 and 17 kDa. In the native polyacrylamide gel these polypeptides migrate as one active enzyme complex, and by gel filtration the peak of enzyme activity can be detected in a position of about 65 kDa. Digestion with endoproteinase Lys-C or endoproteinase Asp-N, followed by peptide analysis with reverse-phase h.p.l.c., reveals an identical peptide pattern for the 24 and 25 kDa bands as well as for the 17 and 18 kDa bands. This treatment further demonstrated a totally different peptide pattern for the 24/25 kDa versus the 17/18 kDa subunit. The N-terminal sequences of the 17 kDa and the 18 kDa peptides were identical, as determined by Edman degradation. The N-termini of the 24 kDa and the 25 kDa peptides were blocked. The enzyme was partly resistant to endoglycosidases H and F, but N-glycosidase F transformed the 24/25 kDa band into one 23 kDa band and the 17/18 kDa band into one 16 kDa band. Also, immunological data obtained with antisera produced against these subunits showed that AGA consists of two non-identical polypeptides.

Comparative gas phase and pulsed liquid phase sequencing on a modified Applied Biosystems 477A sequencer

A simple and inexpensive modification of the Applied Biosystems 477A sequencer, to run in the pulsed liquid-phase and in the gas-phase mode of the Edman chemistry, is described. This modification is especially useful for sequencing samples on polyvinylidene difluoride (PVDF) membranes. Additional carriers are required if a sample on a PVDF membrane is sequenced with the pulsed liquid-phase degradation program of the 477A. In the gas-phase mode no such carriers are needed. This eliminates time-consuming preconditioning sequencer cycles and reduces the sequencer background. In addition, initial coupling yields in the gas-phase mode exceeded those in the pulsed liquid-phase mode, whereas the average repetitive yields were similar. Samples spotted onto glass fiber filters pretreated with polybrene and samples spotted or electroblotted onto PVDF membranes were examined. A number of advantages of the gas-phase mode are presented.

Cloning and sequence analysis of a cDNA for human glycosylasparaginase. A single gene encodes the subunits of this lysosomal amidase

We have isolated a full-length cDNA (HPAsn.6) for human placenta glycosylasparaginase using a 221-bp PCR amplified fragment containing rat liver asparaginase gene sequences. The deduced amino acid sequence from the human clone showed sequence identity to both the alpha and beta subunits of the rat enzyme. The human enzyme is encoded as a 34.6 kDa polypeptide that is post-translationally processed to generate two subunits of approx. 19.5 (alpha) and 15 (beta) kDa. A charge enriched region is present at the predicted site where cleavage occurs. Using polyclonal antibodies against the alpha and beta subunits of rat liver asparaginase, we have shown that the human enzyme is similar in structure to the rat enzyme.